
%matplotlib inline 
%config InlineBackend.figure_formats = ['retina']

# Load libraries
import numpy as np 
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np
import warnings
import statsmodels.api as sm
import statsmodels.formula.api as smf
import sklearn

from statsmodels.stats.outliers_influence import variance_inflation_factor
from scipy.stats import norm
from sklearn.preprocessing import StandardScaler
from sklearn.feature_selection import VarianceThreshold
from scipy import stats
from scipy.stats import chi2_contingency
from sklearn.impute import SimpleImputer
from sklearn.model_selection import train_test_split
from sklearn import preprocessing

warnings.filterwarnings("ignore")
sns.set(rc={"figure.figsize": (20, 15)})
sns.set_style("whitegrid")


# Now it is turn to load our train dataset because in this case I will only perform the initial exploratory data analysis. 

df_train = pd.read_csv("train.csv")

print(f"Train set shape:\n{df_train.shape}\n")

Train set shape:
(891, 12)


# info of each of the variables in our train set
df_train.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 891 entries, 0 to 890
Data columns (total 12 columns):
 #   Column       Non-Null Count  Dtype  
---  ------       --------------  -----  
 0   PassengerId  891 non-null    int64  
 1   Survived     891 non-null    int64  
 2   Pclass       891 non-null    int64  
 3   Name         891 non-null    object 
 4   Sex          891 non-null    object 
 5   Age          714 non-null    float64
 6   SibSp        891 non-null    int64  
 7   Parch        891 non-null    int64  
 8   Ticket       891 non-null    object 
 9   Fare         891 non-null    float64
 10  Cabin        204 non-null    object 
 11  Embarked     889 non-null    object 
dtypes: float64(2), int64(5), object(5)
memory usage: 83.7+ KB


# Lets take a view on thej first data of the set 
df_train.head(10)


# Lets drop the passengerid from our dataset because as we could see is only a number to maintain a counter of the number of passenger
df_train.drop(columns=["PassengerId"], axis=1, inplace=True)
df_train.head(5)


# Lets select columns with numerical data 
df_train_num = df_train.select_dtypes(exclude=["object"])
df_train_num.head()


# Lets drop columns where the values are mainly constants because they are not to offer information about the data 
sel = VarianceThreshold(threshold=0.05)

# fit finds the features with constant variance 
sel.fit(df_train_num.iloc[:, 1:])

# get the number of features that are not constant 
print(f"Number of retained features: {sum(sel.get_support())}")

print(f"\nNumber of quasi_constant features: {len(df_train_num.iloc[:, 1:].columns) - sum(sel.get_support())}")

quasi_constant_features_list = [x for x in df_train_num.iloc[:, 1:].columns if x not in df_train_num.iloc[:, 1:].columns[sel.get_support()]]
print(f"\nQuasi-constant features to be dropped: {quasi_constant_features_list}")

Number of retained features: 5

Number of quasi_constant features: 0

Quasi-constant features to be dropped: []


#Lets plot the distribution of all the numerical data
fig_ = df_train_num.hist(figsize=(8, 10), bins=50, color="deepskyblue",
                         edgecolor="black", xlabelsize=8, ylabelsize=8)


# Lets take a view on the percentage of passengers survived
survived_passengers = df_train.Survived.value_counts()[1]
print(f"\n% Survived passengers: {survived_passengers / len(df_train)}")

% Survived passengers: 0.3838383838383838


# Heatmap for all the remaining numerical data including the target 'Survived'

# This will show us the correlations between variables. I will try to mark the variables which have more than 0.3 of correlation 
pd.options.display.float_format = "{:,.2f}".format

# Define correlation matrix
corr_matrix = df_train_num.corr()

# Replace correlation < |0.20| by 0 for a better visibility
corr_matrix[(corr_matrix < 0.20) & (corr_matrix > -0.20)] = 0

# Mask the upper part of the heatmap
mask = np.triu(np.ones_like(corr_matrix, dtype=bool))

# Choose the color map
cmap = "viridis"

# plot the heatmap
sns.heatmap(corr_matrix, mask=mask, vmax=1.0, vmin=-1.0, linewidths=0.1,
            annot_kws={"size": 9, "color": "black"}, square=True, cmap=cmap, annot=True)

<AxesSubplot:>


# Let's select features where the correlation with 'SalePrice' is higher than |0.2|
# -1 because the latest row is SalePrice
df_num_corr = df_train_num.corr()["Survived"][1:]

# Correlated features (r2 > 0.5)
high_features_list = df_num_corr[abs(
    df_num_corr) >= 0.5].sort_values(ascending=False)
print(f"{len(high_features_list)} strongly correlated values with SalePrice:\n{high_features_list}\n")

# Correlated features (0.2 < r2 < 0.5)
low_features_list = df_num_corr[(abs(df_num_corr) < 0.5) & (
    abs(df_num_corr) >= 0.2)].sort_values(ascending=False)
print(
    f"{len(low_features_list)} slightly correlated values with SalePrice:\n{low_features_list}")

0 strongly correlated values with SalePrice:
Series([], Name: Survived, dtype: float64)

2 slightly correlated values with SalePrice:
Fare      0.26
Pclass   -0.34
Name: Survived, dtype: float64


# Plcass vs Survived
df_train[["Pclass","Survived"]].groupby(["Pclass"], as_index = False).mean().sort_values(by="Survived",ascending = False)


# Features with low correlation (between 0.2 and 0.5)
low_features = df_num_corr[(abs(df_num_corr) >= 0.2) & (
    abs(df_num_corr) < 0.5)].index.tolist()
low_features.append("Survived")
df_low_features = df_train_num.loc[:, low_features]
df_low_features.head(5)

plt.style.use("seaborn-whitegrid")  # define figures style
fig, ax = plt.subplots(round(len(low_features) / 3), 3)

for i, ax in enumerate(fig.axes):
    # plot the correlation of each feature with SalePrice
    if i < len(low_features) - 1:
        sns.regplot(x=low_features[i], y="Survived", data=df_low_features, ax=ax, scatter_kws={
                    "color": "deepskyblue"}, line_kws={"color": "black"},)


sns.pairplot(df_train_num)

<seaborn.axisgrid.PairGrid at 0x2964307dfc0>


survived = 'survived'
not_survived = 'not survived'
fig, axes = plt.subplots(nrows=1, ncols=1,figsize=(6, 6))
ax = sns.distplot(df_train_num[df_train_num['Survived']==1].Age.dropna(), bins=40, label = survived, ax = axes, color="green")
ax = sns.distplot(df_train_num[df_train_num['Survived']==0].Age.dropna(), bins=40, label = not_survived, ax = axes, color="red")
ax.legend()

<matplotlib.legend.Legend at 0x2961cd0e6b0>


child_data = df_train_num[df_train_num['Age'] < 10]

survived_child = len(child_data[child_data["Survived"] == 1])
not_survived_child = len(child_data[child_data["Survived"] == 0])

print(f"\nPercentage of survived child: {survived_child / len(child_data)}")
print(f"\nPercentage of not survived child: {not_survived_child / len(child_data)}")

Percentage of survived child: 0.6129032258064516

Percentage of not survived child: 0.3870967741935484


data = [df_train]
for dataset in data:
    dataset['relatives'] = dataset['SibSp'] + dataset['Parch']
    dataset.loc[dataset['relatives'] > 0, 'travelled_alone'] = 'No'
    dataset.loc[dataset['relatives'] == 0, 'travelled_alone'] = 'Yes'
axes = sns.factorplot('relatives','Survived', 
                      data=df_train, aspect = 2.5, )


# Check the NaN of the train set by ploting percent of missing values per column
column_with_nan = df_train_num.columns[df_train_num.isnull().any()]
column_name = []
percent_nan = []

for i in column_with_nan:
    column_name.append(i)
    percent_nan.append(
        round(df_train_num[i].isnull().sum()*100/len(df_train_num), 2))

tab = pd.DataFrame(column_name, columns=["Column"])
tab["Percent_NaN"] = percent_nan
tab.sort_values(by=["Percent_NaN"], ascending=False, inplace=True)

# Define figure parameters
sns.set(rc={"figure.figsize": (5, 2)})
sns.set_style("whitegrid")

# Plot results
p = sns.barplot(x="Percent_NaN", y="Column", data=tab,
                edgecolor="black", color="deepskyblue")

p.set_title("Percent of NaN per column of the train set\n", fontsize=20)
p.set_xlabel("\nPercent of NaN (%)", fontsize=20)
p.set_ylabel("Column Name\n", fontsize=20)

Text(0, 0.5, 'Column Name\n')


# Lets fill this missing values with a simpleimputer using median values
# # Imputation of missing values (NaNs) with SimpleImputer
my_imputer = SimpleImputer(strategy="median")
df_train_imputed = pd.DataFrame(my_imputer.fit_transform(df_train_num))
df_train_imputed.columns = df_train_num.columns


# Lets check the distribution of each imputed feature before and after imputation

# Define figure parameters
sns.set(rc={"figure.figsize": (9, 4)})
sns.set_style("whitegrid")
fig, axes = plt.subplots(1, 2)

# Plot the results
for feature, fig_pos in zip(["Age"], [0]):

    """Features distribution before and after imputation"""

    # before imputation
    p = sns.histplot(ax=axes[0], x=df_train_num[feature],
                     kde=True, bins=30, color="deepskyblue", edgecolor="black")
    p.set_ylabel(f"Before imputation", fontsize=14)

    # after imputation
    q = sns.histplot(ax=axes[1], x=df_train_imputed[feature],
                     kde=True, bins=30, color="darkorange", edgecolor="black")
    q.set_ylabel(f"After imputation", fontsize=14)


# Let's explore a littl bit the count of any of the categorial values. 

def bar_plot(variable):
    """
        input: variable ex: "Sex"
        output: bar plot & value count
    """
    # get feature
    var = df_train[variable]
    # count number of categorical variable(value/sample)
    varValue = var.value_counts()
    
    # visualize
    plt.figure(figsize = (9,3))
    plt.bar(varValue.index, varValue)
    plt.xticks(varValue.index, varValue.index.values)
    plt.ylabel("Frequency")
    plt.title(variable)
    plt.show()
    print("{}: \n {}".format(variable,varValue))


category1 = ["Survived","Sex","Pclass","Embarked","SibSp", "Parch"]
for c in category1:
    bar_plot(c)

Survived: 
 0    549
1    342
Name: Survived, dtype: int64

Sex: 
 male      577
female    314
Name: Sex, dtype: int64

Pclass: 
 3    491
1    216
2    184
Name: Pclass, dtype: int64

Embarked: 
 S    644
C    168
Q     77
Name: Embarked, dtype: int64

SibSp: 
 0    608
1    209
2     28
4     18
3     16
8      7
5      5
Name: SibSp, dtype: int64

Parch: 
 0    678
1    118
2     80
5      5
3      5
4      4
6      1
Name: Parch, dtype: int64


# Let's analyse the categorical variables with more possibilities in their values

category2 = ["Cabin", "Name", "Ticket"]
for c in category2:
    print("{} \n".format(df_train[c].value_counts()))

B96 B98        4
G6             4
C23 C25 C27    4
C22 C26        3
F33            3
              ..
E34            1
C7             1
C54            1
E36            1
C148           1
Name: Cabin, Length: 147, dtype: int64 

Braund, Mr. Owen Harris                     1
Boulos, Mr. Hanna                           1
Frolicher-Stehli, Mr. Maxmillian            1
Gilinski, Mr. Eliezer                       1
Murdlin, Mr. Joseph                         1
                                           ..
Kelly, Miss. Anna Katherine "Annie Kate"    1
McCoy, Mr. Bernard                          1
Johnson, Mr. William Cahoone Jr             1
Keane, Miss. Nora A                         1
Dooley, Mr. Patrick                         1
Name: Name, Length: 891, dtype: int64 

347082      7
CA. 2343    7
1601        7
3101295     6
CA 2144     6
           ..
9234        1
19988       1
2693        1
PC 17612    1
370376      1
Name: Ticket, Length: 681, dtype: int64


# Categorical to Quantitative relationship
categorical_features = [
    i for i in df_train.columns if df_train.dtypes[i] == "object"]
categorical_features.append("Survived")

# Train set
df_train_categ = df_train[categorical_features]
df_train_categ.head(5)


# Countplot for each of the categorical features in the train set
fig, axes = plt.subplots(round(len(df_train_categ.columns) / 3), 3, figsize=(10, 6))

for i, ax in enumerate(fig.axes):
    # plot barplot of each feature
    if i < len(df_train_categ.columns) - 1:
        ax.set_xticklabels(ax.xaxis.get_majorticklabels(), rotation=45)
        sns.countplot(
            x=df_train_categ.columns[i], alpha=0.7, data=df_train_categ, ax=ax)


# With the boxplot we can see the variation of the target 'Survived' in each of the categorical features
fig, axes = plt.subplots(
    round(len(df_train_categ.columns)/3), 3, figsize=(30, 15))

for i, ax in enumerate(fig.axes):
    # plot the variation of SalePrice in each feature
    if i < len(df_train_categ.columns) - 1:
        ax.set_xticklabels(ax.xaxis.get_majorticklabels(), rotation=45)
        sns.boxplot(
            x=df_train_categ.columns[i], y="Survived", data=df_train_categ, ax=ax)

fig.tight_layout()


# Lets analyse the male or female influence depending on the Age 
survived = 'survived'
not_survived = 'not survived'
fig, axes = plt.subplots(nrows=1, ncols=2,figsize=(16, 8))
women = df_train[df_train['Sex']=='female']
men = df_train[df_train['Sex']=='male']
ax = sns.distplot(women[women['Survived']==1].Age.dropna(), bins=18, label = survived, ax = axes[0], kde =False, color="green")
ax = sns.distplot(women[women['Survived']==0].Age.dropna(), bins=40, label = not_survived, ax = axes[0], kde =False, color="red")
ax.legend()
ax.set_title('Female')
ax = sns.distplot(men[men['Survived']==1].Age.dropna(), bins=18, label = survived, ax = axes[1], kde = False, color="green")
ax = sns.distplot(men[men['Survived']==0].Age.dropna(), bins=40, label = not_survived, ax = axes[1], kde = False, color="red")
ax.legend()
_ = ax.set_title('Male');


# We will do feature engineering and create a new sex that is children, so that we could differenciate between male, female and children. 
print(f"Female before Feature engineering : {len(df_train[df_train['Sex']=='female'])}")
print(f"Male before Feature engineering : {len(df_train[df_train['Sex']=='male'])}")

df_train.loc[df_train['Age'] < 15, 'Sex'] = "Children"

print(f"Female after Feature engineering : {len(df_train[df_train['Sex']=='female'])}")
print(f"Male after Feature engineering : {len(df_train[df_train['Sex']=='male'])}")
print(f"Children after Feature engineering : {len(df_train[df_train['Sex']=='Children'])}")

Female before Feature engineering : 314
Male before Feature engineering : 577
Female after Feature engineering : 275
Male after Feature engineering : 538
Children after Feature engineering : 78


g = sns.catplot(x="Sex", y="Survived", col="Pclass",
                data=df_train, saturation=.5,
                kind="bar", ci=None, aspect=.6)

(g.set_axis_labels("", "Survival Rate")
  .set_xticklabels(["Men", "Women", "Children"])
  .set_titles("{col_name} {col_var}")
  .set(ylim=(0, 1))
  .despine(left=True))

<seaborn.axisgrid.FacetGrid at 0x29682e07c40>


# Lets fill the embarked column missin values with the mode 
embarked_mode = df_train_categ['Embarked'].mode()
data = [df_train_categ]
for dataset in data:
    dataset['Embarked'] = dataset['Embarked'].fillna(embarked_mode)


FacetGrid = sns.FacetGrid(df_train, row='Embarked', size=2, aspect=1.6)
FacetGrid.map(sns.pointplot, 'Pclass', 'Survived', 'Sex', order=None, hue_order=None )
FacetGrid.add_legend()

<seaborn.axisgrid.FacetGrid at 0x2969b42a500>


# Visualize data characteristics
df_train.describe()


list1 = ["SibSp", "Parch", "Age", "Fare", "Survived"]
sns.heatmap(df_train[list1].corr(), annot = True, fmt = ".2f")
plt.show()


g = sns.factorplot(x = "SibSp", y = "Survived", data = df_train, kind = "bar", size = 6)
g.set_ylabels("Survived Probability")
plt.show()


g = sns.factorplot(x = "Parch", y = "Survived", kind = "bar", data = df_train, size = 6)
g.set_ylabels("Survived Probability")
plt.show()


g = sns.factorplot(x = "Pclass", y = "Survived", data = df_train, kind = "bar", size = 6)
g.set_ylabels("Survived Probability")
plt.show()


g = sns.FacetGrid(df_train, col = "Survived")
g.map(sns.distplot, "Age", bins = 25)
plt.show()


g = sns.FacetGrid(df_train, col = "Survived", row = "Pclass", size = 2)
g.map(plt.hist, "Age", bins = 25)
g.add_legend()
plt.show()


g = sns.FacetGrid(df_train, row = "Embarked", col = "Survived", size = 3)
g.map(sns.barplot, "Sex", "Fare")
g.add_legend()
plt.show()


sns.factorplot(x = "Sex", y = "Age", data = df_train, kind = "box")
plt.show()


sns.factorplot(x = "Sex", y = "Age", hue = "Pclass",data = df_train, kind = "box")
plt.show()


sns.factorplot(x = "Parch", y = "Age", data = df_train, kind = "box")
sns.factorplot(x = "SibSp", y = "Age", data = df_train, kind = "box")
plt.show()


df_train["Name"].head(10)

0                              Braund, Mr. Owen Harris
1    Cumings, Mrs. John Bradley (Florence Briggs Th...
2                               Heikkinen, Miss. Laina
3         Futrelle, Mrs. Jacques Heath (Lily May Peel)
4                             Allen, Mr. William Henry
5                                     Moran, Mr. James
6                              McCarthy, Mr. Timothy J
7                       Palsson, Master. Gosta Leonard
8    Johnson, Mrs. Oscar W (Elisabeth Vilhelmina Berg)
9                  Nasser, Mrs. Nicholas (Adele Achem)
Name: Name, dtype: object


name = df_train["Name"]
df_train["Title"] = [i.split(".")[0].split(",")[-1].strip() for i in name]
df_train["Title"].head(10)

0        Mr
1       Mrs
2      Miss
3       Mrs
4        Mr
5        Mr
6        Mr
7    Master
8       Mrs
9       Mrs
Name: Title, dtype: object


sns.countplot(x="Title", data = df_train)
plt.xticks(rotation = 60)
plt.show()


# convert to categorical
df_train["Title"] = df_train["Title"].replace(["Lady","the Countess","Capt","Col","Don","Dr","Major","Rev","Sir","Jonkheer","Dona"],"other")
df_train["Title"] = [0 if i == "Master" else 1 if i == "Miss" or i == "Ms" or i == "Mlle" or i == "Mrs" else 2 if i == "Mr" else 3 for i in df_train["Title"]]
df_train["Title"].head(20)

0     2
1     1
2     1
3     1
4     2
5     2
6     2
7     0
8     1
9     1
10    1
11    1
12    2
13    2
14    1
15    1
16    0
17    2
18    1
19    1
Name: Title, dtype: int64


sns.countplot(x="Title", data = df_train)
plt.xticks(rotation = 60)
plt.show()


g = sns.factorplot(x = "Title", y = "Survived", data = df_train, kind = "bar")
g.set_xticklabels(["Master","Mrs","Mr","Other"])
g.set_ylabels("Survival Probability")
plt.show()


# Drop names column now that we have extracted useful information
df_train.drop(labels = ["Name"], axis = 1, inplace = True)


train_df = pd.get_dummies(df_train,columns=["Title"])
train_df.head()


train_df = pd.get_dummies(train_df, columns=["Embarked"])
train_df.head()


train_df["Ticket"].head(20)

0            A/5 21171
1             PC 17599
2     STON/O2. 3101282
3               113803
4               373450
5               330877
6                17463
7               349909
8               347742
9               237736
10             PP 9549
11              113783
12           A/5. 2151
13              347082
14              350406
15              248706
16              382652
17              244373
18              345763
19                2649
Name: Ticket, dtype: object


a = "A/5. 2151"
a.replace(".","").replace("/","").strip().split(" ")[0]

'A5'


tickets = []
for i in list(train_df.Ticket):
    if not i.isdigit():
        tickets.append(i.replace(".","").replace("/","").strip().split(" ")[0])
    else:
        tickets.append("x")
train_df["Ticket"] = tickets


train_df["Ticket"].head(20)

0         A5
1         PC
2     STONO2
3          x
4          x
5          x
6          x
7          x
8          x
9          x
10        PP
11         x
12        A5
13         x
14         x
15         x
16         x
17         x
18         x
19         x
Name: Ticket, dtype: object


train_df = pd.get_dummies(train_df, columns= ["Ticket"], prefix = "T")
train_df.head(10)


train_df["Pclass"] = train_df["Pclass"].astype("category")
train_df = pd.get_dummies(train_df, columns= ["Pclass"])
train_df.head()


train_df["Sex"] = train_df["Sex"].astype("category")
train_df = pd.get_dummies(train_df, columns=["Sex"])
train_df.head()


# Drop cabin column because it not affect to the target variable 
train_df.drop(labels = ["Cabin"], axis = 1, inplace = True)


male_mean = df_train[df_train['Sex'] == 'male'].Survived.mean()
female_mean = df_train[df_train['Sex'] == 'female'].Survived.mean()
print ("Male survival mean: " + str(male_mean))
print ("female survival mean: " + str(female_mean))

print ("The mean difference between male and female survival rate: " + str(female_mean - male_mean))

Male survival mean: 0.16356877323420074
female survival mean: 0.76
The mean difference between male and female survival rate: 0.5964312267657993


# separating male and female dataframe. 
import random
from statsmodels.stats.weightstats import ztest as ztest
male = df_train[df_train['Sex'] == 'male']
female = df_train[df_train['Sex'] == 'female']

## empty list for storing mean sample
m_mean_samples = []
f_mean_samples = []

for i in range(50):
    m_mean_samples.append(np.mean(random.sample(list(male['Survived']),50,)))
    f_mean_samples.append(np.mean(random.sample(list(female['Survived']),50,)))
    

# Print them out
print (f"Male mean sample mean: {round(np.mean(m_mean_samples),2)}")
print (f"Female mean sample mean: {round(np.mean(f_mean_samples),2)}")
print (f"Difference between male and female mean sample mean: {round(np.mean(f_mean_samples) - np.mean(m_mean_samples),2)}")

Male mean sample mean: 0.16
Female mean sample mean: 0.76
Difference between male and female mean sample mean: 0.6


#perform two sample z-test
ztest(male['Survived'], female['Survived'], value=0.1)

(-24.04918087261463, 8.514134674211484e-128)

	Survived	Pclass	Age	SibSp	Fare
0	0	3	22.00	1	7.25
1	1	1	38.00	1	71.28
2	1	3	26.00	0	7.92
3	1	1	35.00	1	53.10
4	0	3	35.00	0	8.05

	Pclass	Survived
0	1	0.63
1	2	0.47
2	3	0.24

	Survived	Pclass	Age	SibSp	Parch	Fare	relatives
count	891.00	891.00	714.00	891.00	891.00	891.00	891.00
mean	0.38	2.31	29.70	0.52	0.38	32.20	0.90
std	0.49	0.84	14.53	1.10	0.81	49.69	1.61
min	0.00	1.00	0.42	0.00	0.00	0.00	0.00
25%	0.00	2.00	20.12	0.00	0.00	7.91	0.00
50%	0.00	3.00	28.00	0.00	0.00	14.45	0.00
75%	1.00	3.00	38.00	1.00	0.00	31.00	1.00
max	1.00	3.00	80.00	8.00	6.00	512.33	10.00

	Survived	Age	SibSp	Fare	Cabin	relatives	travelled_alone	Title_1	...	T_x	Pclass_1	Pclass_3	Sex_female	Sex_male
0	0	22.00	1	7.25	NaN	1	No	0	...	0	0	1	0	1
1	1	38.00	1	71.28	C85	1	No	1	...	0	1	0	1	0
2	1	26.00	0	7.92	NaN	0	Yes	1	...	0	0	1	1	0
3	1	35.00	1	53.10	C123	1	No	1	...	1	1	0	1	0
4	0	35.00	0	8.05	NaN	0	Yes	0	...	1	0	1	0	1

Exploratory data analysis exercise. Honors project.¶

Pablo Hinojosa Lopez¶

Variable description¶

Explore numerical features¶

Categorical Features¶

Feature Engineering¶

Data visualization¶

Hyphothesis Testing¶

Conclusion¶

	PassengerId	Survived	Pclass	Name	Sex	Age	SibSp	Parch	Ticket	Fare	Cabin	Embarked
0	1	0	3	Braund, Mr. Owen Harris	male	22.00	1	0	A/5 21171	7.25	NaN	S
1	2	1	1	Cumings, Mrs. John Bradley (Florence Briggs Th...	female	38.00	1	0	PC 17599	71.28	C85	C
2	3	1	3	Heikkinen, Miss. Laina	female	26.00	0	0	STON/O2. 3101282	7.92	NaN	S
3	4	1	1	Futrelle, Mrs. Jacques Heath (Lily May Peel)	female	35.00	1	0	113803	53.10	C123	S
4	5	0	3	Allen, Mr. William Henry	male	35.00	0	0	373450	8.05	NaN	S
5	6	0	3	Moran, Mr. James	male	NaN	0	0	330877	8.46	NaN	Q
6	7	0	1	McCarthy, Mr. Timothy J	male	54.00	0	0	17463	51.86	E46	S
7	8	0	3	Palsson, Master. Gosta Leonard	male	2.00	3	1	349909	21.07	NaN	S
8	9	1	3	Johnson, Mrs. Oscar W (Elisabeth Vilhelmina Berg)	female	27.00	0	2	347742	11.13	NaN	S
9	10	1	2	Nasser, Mrs. Nicholas (Adele Achem)	female	14.00	1	0	237736	30.07	NaN	C